[antlr-interest] Request for a feature (if possible in V4)

[Gokulakannan Somasundaram]

Hi All,
        I don't know, whether this is possible. But i just thought of
putting forward my thought. Is it possible to write a parser-mode, where in
it would generate code that is auto left factored. There are several
advantages.
a) When the parser becomes like in my case (around 1000 rules), then parser
generation takes more time, if i increase the value of k.
b) As i increase the value of k, the heap size requirement increases almost
exponentially
c) Runtime performance will definitely increase
d) Parser code size will definitely come down.
     If this is not possible, can someone explain me on the difficulties
encountered in creating such a parser utility?


Thanks,
Gokul.